JP2018058295A - Image processing apparatus, image processing method, and program - Google Patents

Abstract

An object of the present invention is to efficiently perform parallel drawing processing and speed up drawing processing while suppressing an increase in cost due to an increase in hardware resources. A plurality of drawing means for performing drawing processing in parallel with respect to each of a plurality of band-by-band intermediate data corresponding to each of a plurality of band regions, wherein the band region corresponds to a predetermined block region. Multiple drawing units with a large number of pixels in the main scanning direction and a small number of pixels in the sub-scanning direction, and image data in band units generated from intermediate data when the storage capacity is insufficient in the drawing process Each of the plurality of drawing means combines the saved image data read from the storage means with the image data generated by performing a new drawing process on the intermediate data. The raster format image data is generated. [Selection] Figure 4

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program for executing drawing processing in parallel by a plurality of drawing processing units.

  Conventionally, it is widely known that the processing efficiency of image processing can be improved by interpreting drawing data in page units, generating intermediate data called display lists (DL), dividing into rectangular areas in block units and performing parallel processing. It has been. In particular, the block unit of a square tile shape (for example, 32 pixels × 32 pixels) requires only a relatively small memory capacity for primary storage, does not depend on the image size or resolution, and changes in image shape before and after the rotation process. It is excellent in that it does not.

  Further, Patent Document 1 discloses a technique for dividing the intermediate data (DL) into band units including rectangular areas in block units and performing parallel processing after dividing the data into block units in order to distribute the processing load. ing.

  In addition, when the intermediate data (DL) is drawn, if the memory capacity is insufficient, a degeneration operation (fallback) is performed. In the reduction operation, intermediate data (DL) drawing processing is performed in a plurality of times in the memory capacity while temporarily generating image data in the memory.

JP 2012-254583 A

  However, according to the technique of Patent Document 1, if the drawing processing in units of bands is performed in parallel, the memory for storing the image data to be saved at the time of the reduction operation is required for the number of parallel operations of the drawing processing. In the technique disclosed in Patent Document 1, as many decompression processing units as necessary to decompress the saved image data are required for the number of parallel operations of the drawing process. As described above, the conventional parallel drawing processing in units of bands increases the hardware resources (circuit scale), leading to an increase in cost.

  Therefore, even when drawing processing in units of bands is performed in parallel, it is desired to efficiently perform drawing processing and speed up drawing processing while suppressing an increase in cost due to an increase in hardware resources.

  The image processing apparatus according to the present invention is a plurality of drawing means for performing drawing processing in parallel on each of a plurality of intermediate data in units of bands corresponding to each of a plurality of band regions, wherein the band regions are predetermined A plurality of drawing means having a large number of pixels in the main scanning direction and a small number of pixels in the sub-scanning direction relative to the block area, and generated from the intermediate data when storage capacity is insufficient in the drawing process. Storage means for saving and storing image data in units of bands, and each of the plurality of drawing means newly writes the saved image data read from the storage means to the intermediate data. The image data in the raster format is generated by synthesizing with the image data generated by processing.

  According to the present invention, it is possible to efficiently perform parallel drawing processing and speed up drawing processing while suppressing an increase in cost due to an increase in hardware resources.

1 is a block diagram illustrating a configuration of an image forming apparatus. 3 is a diagram illustrating a data flow of PDL printing processing in the image forming apparatus. FIG. FIG. 4 is a diagram illustrating an example of a coordinate system of a rectangular area in a page handled by the image forming apparatus. It is a figure which shows the block structure in connection with a drawing process. It is a figure showing the state where a part of block composition concerning drawing processing stopped. It is a flowchart of a PDL printing process. It is a flowchart of a band height setting process. It is a flowchart of a parallel drawing process. It is a figure explaining the access control of the write of a block unit, and the read of a band unit. It is a flowchart of a band height setting process. It is a figure which shows the positional relationship of the rectangular area of a block unit and a band unit.

  Embodiments of the present invention will be described below in detail with reference to the drawings. In addition, embodiment described below is only an example and does not limit this invention. Moreover, the same code | symbol shows the same thing throughout the drawing.

  FIG. 1 is a block diagram illustrating a configuration example of an image forming apparatus according to an embodiment of the present invention. As illustrated in FIG. 1, the image forming apparatus 100 includes a controller 101, a network 102, an operation unit 103, a reading unit 104, and a printing unit 105.

  The network 102 is a communication unit that is realized by a LAN, a WAN (public line), or the like, and transmits and receives image data and device information between an external apparatus such as a host computer and a server and the image forming apparatus 100. The operation unit 103 is a processing unit that acquires an operation by a user as image processing control information or displays image processing control information to the user. A reading unit (scanner engine) 104 is an image input device, and is a processing unit that captures image data into the image forming apparatus 100 using an optical sensor or the like. A printing unit (printer engine) 105 is an image output device, and is a processing unit that prints image data inside the image forming apparatus 100 on a recording medium.

  The controller 101 is connected to the network 102, the operation unit 103, the reading unit 104, and the printing unit 105, and is a control unit that controls the entire image forming apparatus 100. The controller 101 includes a system bus 106, a communication unit I / F 107, a CPU 108, a RAM 109, a ROM 110, an HDD 111, an operation unit I / F 112, and a reading unit I / F 113. The controller 101 also includes a drawing processing unit 114, a compression processing unit 115, an expansion processing unit 116, a band RAM 117, and a printing unit I / F 118. The controller 101 can be implemented as an image processing apparatus, and may be configured without including the reading unit 104 and the printing unit 105, for example.

  The system bus 106 is a processing unit for connecting the processing units constituting the controller 101 and transmitting / receiving image data and control information between the processing units. The communication unit I / F 107 is realized by, for example, a LAN card or the like, and is an interface unit for transmitting and receiving image data and device information between an external apparatus such as a host computer and a server and the image forming apparatus 100 via the network 102. is there.

  The CPU 108 is a processing unit that controls the entire image forming apparatus 100. In particular, in PDL (page description language) printing processing, rendering data such as PDL data received from an external device via the network 102 is interpreted and converted into intermediate data called DL (display list). Here, description will be made assuming that PDL data is processed as an example of drawing data.

  A RAM (volatile memory) 109 is a storage unit that is used as a work area for the CPU 108 to operate on the system or used as a buffer area for temporarily storing image data. A ROM (nonvolatile memory) 110 is a primary storage unit in which a program for the CPU 108 to start the system is stored. This program is expanded in the RAM 109 when the image forming apparatus 100 is started up and executed by the CPU 108.

  An HDD (Hard Disk Drive) 111 is a large capacity storage unit for storing image data inside the image forming apparatus 100.

  The operation unit I / F 112 is an interface unit that is connected to the operation unit 103 and acquires an operation performed by the user as image processing control information or displays the image processing control information to the user. A reading unit I / F (scanned image processing unit) 113 is connected to a reading unit (scanner engine) 104 for correcting the image data input from the reading unit 104 according to the device characteristics of the reading unit 104. This is an image processing unit for performing the image processing.

  The rendering processing unit (RIP) 114 refers to the vector format intermediate data (DL) generated by the CPU 108, generates raster format image data, and stores it in the RAM 109. In addition, when processing the remaining intermediate data (DL) following the reduction processing, the drawing processing unit 114 generates image data in a raster format by combining the image data saved in the band RAM 117 with the image.

  The compression processing unit 115 performs compression processing on the raster format image data generated by the drawing processing unit 114, generates compressed image data, and stores the compressed image data in the RAM 109.

  The decompression processing unit 116 performs decompression processing of the compressed image data that the compression processing unit 115 performs the compression processing. In particular, in the intermediate data (DL) drawing process following the reduction process, the saved image data is expanded and provided to the drawing processing unit 114 via the band RAM 117.

  The band RAM 117 is a local memory that primarily holds raster-format image data output from the decompression processing unit 116 to the rendering processing unit 114, and is a fallback memory used for rendering processing subsequent to the reduction processing. The band RAM 117 includes an access detection device (not shown), and detects writing of image data to its own specific address and reading of image data from the specific address.

  The operations of the drawing processing unit 114, the compression processing unit 115, the decompression processing unit 116, and the band RAM 117 related to the drawing processing during PDL printing will be described in more detail with reference to FIG.

  A printing unit I / F (print image processing unit) 118 is connected to the printing unit (printer engine) 105, performs image processing for correction in accordance with the device characteristics of the printing unit 105, and then performs printing on the printing unit 105. An image processing unit for outputting image data.

  Although FIG. 1 shows a configuration example in which the drawing processing unit 114 and the expansion processing unit 116 are directly connected via the band RAM 117 as a local memory, the configuration of the image forming apparatus 100 is not limited to this. Absent. Further, for example, the compression processing unit 115 may be configured to be connected on the system bus 106 for general purposes, or connected exclusively to a processing unit that requires compression processing, such as the reading unit I / F 113 or the drawing processing unit 114. May be configured. Similarly, for example, the expansion processing unit 116 may be configured to be connected to the system bus 106 for general purposes, or may be configured to be connected exclusively to a processing unit such as the printing unit I / F 118 that requires expansion processing. May be.

  Next, the relationship between the data flow of the PDL printing process and the image data unit in the image forming apparatus 100 will be described with reference to FIGS. FIG. 2 shows a data flow of PDL printing processing in the image forming apparatus 100. In FIG. 2, particularly, the broken line portion relates to a degeneration operation (fallback) performed when the intermediate data (DL) is larger than a predetermined storage capacity secured on the RAM 109. FIG. 3 shows an example of a coordinate system of a rectangular area in a page handled by the image forming apparatus 100. Note that the data flow shown in FIG. 2 is implemented by the CPU 108 executing a program developed in the RAM 109 of the image forming apparatus 100.

  First, the normal data flow in the PDL printing process will be described with reference to FIG. The “normal time” data flow means a data flow when the degeneration operation (fallback) is not performed.

  As shown in FIG. 2, the image forming apparatus 100 first receives the PDL data transmitted from the host computer by the communication unit I / F 107 and stores it in the RAM 109 or the HDD 111 as the PDL data 201 in units of pages.

  Next, the CPU 108 interprets the PDL data 201 in page units, generates intermediate data (DL) 202 in page units or band units, and stores it in the RAM 109 again. As shown in FIGS. 3A to 3C, the band unit divides a single page in the sub-scanning direction (Y direction), thereby sub-scanning the number of pixels in the main scanning direction (X direction). It means a strip-like rectangular area (band area) having a smaller number of pixels in the direction. In addition, as shown in FIGS. 3A to 3C, the block unit is a rectangular area having a square tile shape in which the number of pixels in the main scanning direction and the number of pixels in the sub-scanning direction are predetermined in the image forming apparatus 100. (Block area). The difference between FIG. 3A, FIG. 3B, and FIG. 3C is whether or not the rectangular area in block units is in a positional relationship included in the rectangular area in band units. FIG. 3A and FIG. 3B show a conventional relationship in which blocks are included in a single band, and the single band is composed of a set of a plurality of blocks. In particular, FIG. 3B shows an example in which the height of the block and the band in the sub-scanning direction (the number of pixels in the sub-scanning direction) is equal. On the other hand, FIG. 3C shows an image data unit suitable for executing drawing processing in parallel in one embodiment of the present invention. In FIG. 3C, the block and the band are not included in each other, and the relationship in which the number of pixels (the height of the block) in the sub-scanning direction of the block is divided by the height of the plurality of bands is satisfied. Specifically, as shown in FIG. 3C, in the single page 1, the number of pixels in the sub-scanning direction of the single block 11 is divided by the height of each band 11 and band 12. Satisfy the relationship. In the following description, examples of the coordinate system of the band unit and the block unit in the page shown in FIGS. 3A to 3C will be used as necessary.

  The CPU 108 also generates intermediate data (DL) and stores it in the RAM 109, and sets parallel drawing processing setting information 203 related to the parallel operation of the drawing processing unit 114 in the internal register of the drawing processing unit 114. The parallel drawing processing setting information 203 includes the start address of the address of the RAM 109 in which intermediate data (DL) is stored, the number of parallel operations of the drawing processing unit 114, and the number of pixels in the sub-scanning direction of the band to be drawn (band height). ) Etc. are included.

  Next, the drawing processing unit 114 refers to the intermediate data (DL) 202 stored in the RAM 109 and executes drawing processing on a plurality of band-by-band rectangular areas in a single page in parallel. The drawing processing unit 114 performs drawing processing in parallel in units of bands, thereby generating raster format image data 204 of a plurality of bands, and temporarily stores them in the RAM 109.

  Next, the compression processing unit 115 reads block-unit image data from the band-unit image data 204 primarily stored in the RAM 109. When a single block is composed of a plurality of bands, a plurality of band-unit image data 204 constituting a single block are combined, and a plurality of band-unit image data 204 is combined to form block-unit image data. Is read. Then, the compression processing unit 115 performs compression processing such as JPEG on the read block-unit image data to generate compressed image data 205 and stores the compressed image data 205 in the RAM 109 or the HDD 111 again.

  Next, the decompression processing unit 116 performs decompression processing such as JPEG on the compressed image data 205 compressed in units of blocks to generate raster format image data. The decompression processing unit 116 combines all block-unit raster format image data constituting a single page, generates page-unit raster format image data 206, and temporarily stores it in the RAM 109 again.

  Next, the printing unit I / F 118 reads the page-unit image data 206 temporarily stored in the RAM 109, performs image processing for correction in accordance with the device characteristics of the printing unit 105, and then performs printing on the printing unit 105. To output the corrected image data.

  Next, the printing unit (printer engine) 105 prints the image data output from the printing unit I / F 118 on a recording medium.

  The normal data flow in the PDL printing process has been described above. Next, the data flow during the degeneration operation (fallback) in the PDL printing process will be described with particular attention to the broken line portion in FIG.

  When the CPU 108 interprets the PDL data 201 and generates the intermediate data (DL) 202, is the intermediate data (DL) 202 constituting a single page larger than a predetermined storage capacity secured on the RAM 109? Judge whether. When the storage capacity is larger than the predetermined storage capacity (that is, when the storage capacity is insufficient), the CPU 108 sequentially stores the RAM 109 as the intermediate data (DL) 202 from the drawing object drawn on the back surface within the range not exceeding the predetermined storage capacity. To remember. Rendering processing is performed on the intermediate data (DL) 202 by the rendering processing unit 114, and raster format image data 207 obtained through the compression processing unit 115 and the decompression processing unit 116 is temporarily stored in the band RAM 117. . This image data 207 is referred to as back image data in the degeneration operation (fallback). The back image data is temporarily stored in the band RAM 117 as image data for image synthesis, and the drawing processing unit 114 refers to this and uses it for the drawing processing. That is, the drawing processing unit 114 obtains a final drawing result by synthesizing the drawing result of the remaining intermediate data (DL) that did not fit in the RAM 109 and the back image acquired from the band RAM 117. Can do. In this way, by repeating until the intermediate data (DL) 202 fits in a predetermined storage capacity on the RAM 109 in accordance with the above-described degeneration operation (fallback), it is limited by limited hardware resources constituting the image forming apparatus 100. The drawing process can be realized without any problem.

  Next, with reference to FIG. 4, the operation of the drawing process during PDL printing in one embodiment of the present invention will be described. FIG. 4 illustrates a drawing process performed in parallel on a band basis using intermediate data (DL) 202 and compressed image data 205 that are primarily stored in the RAM 109 during a reduction operation (fallback), and the obtained image data 204 is stored in the RAM 109. The operation to memorize is shown.

  FIG. 4A shows the operation of the drawing process when the rectangular area in units of blocks is included in the rectangular area in units of bands as shown in FIG. In FIG. 4A, the image forming apparatus 100 uses a first drawing processing unit 401 and a second drawing processing unit 402 constituting the drawing processing unit 114, and a plurality of images in a single page of the intermediate data (DL) 202. Each band is drawn in parallel. The first drawing processing unit 401 performs drawing processing on odd-numbered bands in a single page, skips even-numbered bands, and updates the position coordinates for each scan line of the drawing object. The second drawing processing unit 402 draws even-numbered bands in a single page, skips odd-numbered bands, and updates the position coordinates for each scan line of the drawing object. As described above, since the first drawing processing unit 401 and the second drawing processing unit 402 each acquire a drawing object to be drawn in a band unit, the intermediate data (DL) 202 temporarily stored in the RAM 109 is a page. Units or band units may be used.

  The first decompression processing unit 421 and the second decompression processing unit 422 constituting the decompression processing unit 116 operate in cooperation with the first rendering processing unit 401 and the second rendering processing unit 402 constituting the rendering processing unit 114, respectively. To do. The first decompression processing unit 421 performs the decompression process on the block 1 image data corresponding to the band 1 and the band 3 shown in FIG. 3B acquired from the compressed image data 205 in order, and configures the band RAM 117. To the first band RAM 411. Further, the second decompression processing unit 422 performs decompression processing on the block-unit image data corresponding to the band 2 and the band 4 shown in FIG. Is output to the second band RAM 412.

  In response to this, the first drawing processing unit 401 stores the band-unit image data composed of a plurality of blocks corresponding to the band 1 and the band 3 shown in FIG. The images are read in order, and the images are combined and transferred to the RAM 109 during the drawing process. Similarly, the second drawing processing unit 402 sequentially reads out image data in band units composed of a plurality of blocks corresponding to band 2 and band 4 shown in FIG. The images are combined and transferred to the RAM 109 during the drawing process.

  The first band RAM 411 includes a double buffer that can simultaneously execute writing by the first decompression processing unit 421 and reading by the first drawing processing unit 401. Therefore, for example, the writing process of the blocks 31 to 34 corresponding to the band 3 by the first decompression processing unit 421 and the reading process of the band 1 by the first drawing processing unit 401 can be executed simultaneously. Similarly, the second band RAM 412 is composed of a double buffer capable of simultaneously executing writing by the second decompression processing unit 422 and reading by the second drawing processing unit 402. Therefore, for example, the writing process of the blocks 41 to 44 corresponding to the band 4 by the second decompression processing unit 422 and the reading process of the band 2 by the second drawing processing unit 402 can be executed simultaneously.

  As described above, in the example of FIG. 4A, not only the hardware resources necessary for parallelizing the drawing processing unit 114 but also the expansion processing unit 116 and the band RAM 117 are the number of parallel operations of the drawing processing unit 114. Only minutes are needed.

  In FIG. 4B, as shown in FIG. 3C, the block-unit rectangular area is not included in the band-unit rectangular area, and the number of pixels in the sub-scanning direction of the block unit is calculated in parallel by the drawing processing unit 114. An operation of a drawing process when divided by a plurality of band-based heights (number of pixels) based on the number of operations is shown. In FIG. 4B, the image forming apparatus 100 draws a plurality of bands in a single page in parallel using the first drawing processing unit 401 and the second drawing processing unit 402 constituting the drawing processing unit 114. Process. The first drawing processing unit 401 performs drawing processing on odd-numbered bands in a single page, skips even-numbered bands, and updates the position coordinates for each scan line of the drawing object. The second drawing processing unit 402 draws even-numbered bands in a single page, skips odd-numbered bands, and updates the position coordinates for each scan line of the drawing object. As described above, since the first drawing processing unit 401 and the second drawing processing unit 402 each acquire a drawing object to be drawn in a band unit, the intermediate data (DL) 202 temporarily stored in the RAM 109 is a page. Units or band units may be used.

  The decompression processing unit 116 in FIG. 4B includes a single decompression processing unit 421 that operates in cooperation with both the first drawing processing unit 401 and the second drawing processing unit 402. The band RAM 117 is composed of a single band RAM 411.

  A single decompression processing unit 421 constituting the decompression processing unit 116 sequentially decompresses image data in block units corresponding to the bands 11 and 12 shown in FIG. 3C acquired from the compressed image data 205. Processing is performed and output to the band RAM 411.

  In response to this, the first drawing processing unit 401 stores the image data in units of bands corresponding to the band 11 and the band 21 shown in FIG. To the RAM 109. Similarly, the second rendering processing unit 402 is primarily stored in the band RAM 411, sequentially reads out band-unit image data corresponding to the bands 12 and 22 shown in FIG. 3C, and combines the images during the rendering process. Transfer to the RAM 109.

  The band RAM 411 includes a double buffer that can simultaneously execute writing by the expansion processing unit 421 and reading by the first drawing processing unit 401 and the second drawing processing unit 402. Therefore, for example, the writing process of the blocks 21 to 24 corresponding to the band 21 and the band 22 by the expansion processing unit 421 and the reading process of the band 11 by the first drawing processing unit 401 can be executed simultaneously. Similarly, the writing processing of the blocks 21 to 24 corresponding to the band 21 and the band 22 by the expansion processing unit 421 and the reading processing of the band 12 by the second drawing processing unit 402 can be executed simultaneously.

  As described above, in the example of FIG. 4B, even if the drawing processing units 114 are parallelized, it is sufficient that the drawing processing unit 114 has as many hardware resources as the number of parallel operations. And it is not necessary to add the hardware resource of band RAM117.

  5 changes the parallel drawing processing setting information of the drawing processing unit 114 in the image forming apparatus 100 in FIG. 4B, stops the second drawing processing unit 402, and operates only the first drawing processing unit 401. The operation of the drawing process in the case of being made is shown. In this case, since the number of parallel operations of the drawing processing unit 114 is 1, the positional relationship is the same as in FIG. 3B in which the rectangular area in units of blocks is included in the rectangular area in units of bands.

  In FIG. 5, a single decompression processing unit 421 performs decompression processing on the block unit image data corresponding to band 1 and band 2 shown in FIG. To the band RAM 411.

  In response to this, only the first drawing processing unit 401 sequentially reads out band-unit image data composed of a plurality of blocks corresponding to band 1 and band 2 that are primarily stored in the band RAM 411, and synthesizes the images during the drawing process. To the RAM 109. In this way, a single page is sequentially drawn in band units.

  The band RAM 411 is composed of a double buffer that can simultaneously execute writing by the expansion processing unit 421 and reading by the drawing processing unit 401. Therefore, for example, the writing processing of the blocks 21 to 24 in the band 2 by the expansion processing unit 421 and the reading processing of the band 1 by the drawing processing unit 401 can be executed simultaneously.

  As described above, in the example of FIG. 5, even when the parallel drawing processing setting information of the drawing processing unit 114 is changed and some of the plurality of drawing processing units are stopped, By making the band heights correspond to each other, it is possible to realize the same drawing process as in parallel operation.

  FIG. 6 shows a flowchart of the PDL printing process in the image forming apparatus 100. The processing shown in the flowchart of FIG. 6 is implemented by the CPU 108 executing a program developed in the RAM 109 of the image forming apparatus 100.

  First, in step S601, the CPU 108 receives the PDL data transmitted from the host computer via the communication unit I / F 107 and stores it in the RAM 109 or the HDD 111 as PDL data in units of pages.

  In step S602, the CPU 108 interprets the stored page unit PDL data, and generates band unit intermediate data (DL) to be rendered based on the interpreted information.

  Next, in step S <b> 603, the CPU 108 determines whether there is back image data for image composition used for the degeneration operation (fallback) when the generated intermediate data (DL) is larger than a predetermined storage capacity secured on the RAM 109. judge. If there is back image data for image composition, the process advances to step S604, and the CPU 108 uses the decompression processing unit 116 to perform decompression processing of the back image data. In step S604, the decompression processing unit 116 primarily stores in the band RAM 117 image data obtained by performing decompression processing on the compressed image data in units of blocks as described with reference to FIGS. Accordingly, the decompression processing unit 116 decompresses the compressed image data by the band necessary for the drawing processing unit 114 in the subsequent step S606. On the other hand, if there is no back image for image composition, step S604 is skipped and the process proceeds to step S605.

  Next, in step S <b> 605, the CPU 108 sets the band height in the internal register of the drawing processing unit 114 as parallel drawing processing setting information. The band height setting process in step S605 will be described in detail separately with reference to the flowchart of FIG.

  Next, in step S606, the CPU 108 controls the drawing processing unit 114 to execute drawing processing in parallel in units of bands according to the band height set in step S605, and temporarily stores the image data of the drawing processing result in the RAM 109. To do. The parallel drawing process in step S606 will be described in detail separately with reference to the flowchart of FIG.

  Next, in step S <b> 607, the CPU 108 controls the compression processing unit 115 to execute compression processing, compresses the image data of the rendering processing result temporarily stored in the RAM 109, converts it into compressed image data, and again stores it in the RAM 109. Remember.

  In step S608, the CPU 108 determines whether there is an undrawn object that does not fit in the RAM 109 based on the intermediate data (DL) generated based on the PDL data received in step S601. If there is an undrawn object, the process returns to step S602, and the CPU 108 repeats the process until there is no undrawn object as a degeneration operation (fallback). On the other hand, if there is no undrawn object, the process advances to step S609, and the CPU 108 controls the decompression processing unit 116 to perform decompression processing, decompresses the compressed image data temporarily stored in the RAM 109, and stores the decompressed image data in the RAM 109 again. .

  Next, in step S <b> 610, the CPU 108 controls the printing unit I / F 118 to read the image data stored in the RAM 109, performs image processing for correction in accordance with the device characteristics of the printing unit 105, and then performs printing. The image data is transferred to the unit 105.

  Thereafter, in step S611, the printing unit 105 prints the image data transferred from the printing unit I / F 118 on a recording medium.

  As described above, the PDL printing process by the image forming apparatus 100 is performed. Next, the band height setting process in step S605 described above will be described.

  FIG. 7 shows a flowchart of the band height setting process in step S605 described above. The processing shown in the flowchart of FIG. 7 is implemented by the CPU 108 executing a program developed in the RAM 109 of the image forming apparatus 100.

  First, in step S701, as in step S603, the CPU 108 determines whether there is back image data for image synthesis that is used during a degeneration operation (fallback). If there is no back image data, the process proceeds to step S702, and if there is back image data, the process proceeds to step S703.

  When there is no back image data, the cooperative operation using the band RAM 117 between the drawing processing unit 114 and the expansion processing unit 116 described with reference to FIG. 4 is not performed. Therefore, in step S <b> 702, the CPU 108 determines the band height to an arbitrary size according to the storage capacity that can be secured in the RAM 109, and sets the band height in the internal register of the drawing processing unit 114. Arbitrary size means that it may be set to any band height shown in FIGS. 3A to 3C, for example.

  On the other hand, if there is back image data, the CPU 108 acquires a block unit as a data unit handled in the image forming apparatus 100 in step S703. The block unit means the height and width of the rectangular area. Here, for simplicity, the block unit is described as a square tile shape of 8 pixels (height) × 8 pixels (width). Therefore, the height of the block is 8 pixels.

  In step S <b> 704, the CPU 108 acquires the number of parallel operations of the drawing processing unit 114 that performs drawing processing in units of bands. For example, if the drawing processing unit 114 is in the operation state of FIG. 4B, the number of parallel operations is 2, and if the drawing processing unit 114 is in the operation state of FIG. To get it.

  Next, in step S705, the CPU 108 determines whether or not the number of parallel operations acquired in step S704 is 2 or more. If the number of parallel operations is 1 or less, the process proceeds to step S706. If the number of parallel operations is 2 or more, the process proceeds to step S707.

  When the number of parallel operations is 1 or less, in step S706, the CPU 108 directly sets the height of the block acquired in step S703 as the height of the band processed by the drawing processing unit 114. For example, in the case of the operation state of FIG. 5, since the number of parallel operations is 1 for a block unit of 8 pixels × 8 pixels, the height of the band is calculated as 8 (= 8 ÷ 1). 114 is set in the internal register.

  On the other hand, when the number of parallel operations is 2 or more, in step S707, the CPU 108 equally divides the height of the block acquired in step S703 by the number of parallel operations acquired in step S704. For example, in the case of the operation state of FIG. 4B, the number of parallel operations is 2 for a block unit of 8 pixels × 8 pixels, so the band height is calculated as 4 (= 8 ÷ 2).

  In step S708, the value calculated by equal division is set as the band height in the internal register of the drawing processing unit 114.

  As described above, the height of the band is set based on the presence / absence of the back image data and the number of parallel operations of the drawing processing unit. Next, the parallel drawing process in step S606 described above will be described.

  FIG. 8 shows a flowchart of the parallel drawing process in step S606 described above. The processing shown in the flowchart of FIG. 8 is performed by the drawing processing unit 114 when the CPU 108 executes a program developed in the RAM 109 of the image forming apparatus 100. Hereinafter, description will be made assuming that the drawing processing unit 114 operates based on the control of the CPU 108.

  First, in step S <b> 801, the read processing unit of each drawing processing unit that operates in parallel within the drawing processing unit 114 acquires the start address of the intermediate data (DL) set by the CPU 108 in the internal register of the drawing processing unit 114. .

  Next, in step S802, each drawing processing unit of the drawing processing unit 114 acquires the band height set in the internal register of the drawing processing unit 114 in step S605.

  Next, in step S <b> 803, each drawing processing unit of the drawing processing unit 114 acquires the number of parallel operations for executing drawing processing in units of bands inside the drawing processing unit 114 and its own ID from the internal register. For example, in the case of the first drawing processing unit 401 of the drawing processing unit 114 shown in FIG. 4B, the number of parallel operations is 2, and the ID of 1 is acquired. In the case of the second drawing processing unit 402, 2 is acquired as the number of parallel operations, and 2 is acquired as its own ID.

  Next, in step S804, the drawing processing units 401 and 402 of the drawing processing unit 114 acquire a drawing object included in the processing target band from the intermediate data (DL).

  Next, in step S805, each drawing processing unit of the drawing processing unit 114 collates the band number of the processing target band with its own ID, and whether the band is a drawing processing target band for which the drawing process is performed. Determine whether or not. That is, for example, in the case of the first drawing processing unit 401 shown in FIG. 4B, since its own ID is 1, the odd-numbered bands such as the band 11 and the band 21 shown in FIG. Are determined to be their own drawing processing target bands. In the case of the second drawing processing unit 402 shown in FIG. 4B, since its own ID is 2, even-numbered bands such as the band 12 and the band 22 shown in FIG. It is determined that the band is a drawing processing target band. If the band to be processed is the band to be drawn, the process proceeds to step S806. If the band to be processed is not the band to be drawn, the process proceeds to step S809.

  In step S <b> 806, each drawing processing unit of the drawing processing unit 114 determines whether there is back image data for image composition used for the degeneration operation (fallback), as in step S <b> 603 described above. If there is back image data, the process proceeds to step S807, and the back image data corresponding to the band to be processed is acquired from the band RAM 117. Here, as described with reference to FIG. 5, the back surface image data generated by the expansion processing in step S604 described above is temporarily stored in the band RAM 117 for the band necessary for drawing the band area to be processed. Yes. On the other hand, when there is no back image data, step S807 is skipped and the process proceeds to step S808.

  In step S <b> 808, each drawing processing unit of the drawing processing unit 114 acquires a drawing object necessary for drawing the band area to be processed from the intermediate data (DL), and outputs image data as a drawing processing result to the RAM 109. To do. In particular, if there is back image data acquired in step S807, image synthesis processing is performed with reference to the back image data, and image data as a drawing processing result is generated.

  On the other hand, in step S809, each drawing processing unit of the drawing processing unit 114 updates only the position coordinates for each scan line of the drawing object included in the processing target band area, and performs intermediate processing without performing the band area drawing process. A process of skipping data (DL) is performed.

  Next, in step S810, each drawing processing unit of the drawing processing unit 114 determines whether or not there is a next band to be drawn in the intermediate data (DL) stored in the RAM 109. If there is a next band, the process returns to step S804 to repeat the process. If there is no next band, the drawing process for the currently processed page is terminated.

  As described above, the drawing processing unit 114 performs parallel drawing processing.

  FIG. 9 is a diagram for explaining access control for writing (writing) in units of blocks and reading (reading) in units of bands. FIG. 9A shows image data written in block units in the first band RAM 411 of the band RAM 117. FIG. 9B shows image data read from the band RAM 117 in units of bands.

  As illustrated in FIG. 9A, the blocks 11 to 14 are sequentially written to the first band RAM 411 by the first expansion processing unit 421 of the expansion processing unit 116. On the other hand, as shown in FIG. 9B, the band 11 is read from the band RAM 117 by the first drawing processing unit 401 of the drawing processing unit 114, and the band 12 is read from the second drawing processing unit 402 of the drawing processing unit 114. Is read from the band RAM 117.

  The first band RAM 411 draws image data to the address (pixel position) indicated as the write access detection in FIG. 9A as a trigger to operate the decompression processing unit 116 and the drawing processing unit 114 in cooperation with each other. Controls the start of reading (reading) of the processing unit 114. This is because it can be determined that the writing of the band-unit image data has been completed by detecting that the block-unit image data has been written up to the address (pixel position).

  Similarly, the first band RAM 411 performs the write operation of the decompression processing unit 116 using the reading of image data from all addresses (pixel positions) indicated as read access detection in FIG. (Export) Controls the start. By detecting that the image data in the band unit has been read up to all the addresses (pixel positions), it can be determined that the reading of the image data in all the band units including the image data in the block unit has been completed. It is. This means that the first band RAM 411 has a free area for new writing.

  As described above, according to the embodiment of the present invention, the first drawing processing unit 401 and the second drawing processing unit 401 and the second drawing unit 116 and the band RAM 117 are not expanded by the number of parallel operations of the drawing processing unit. The parallel operation of the drawing processing unit 402 can be realized. That is, it is possible to efficiently perform parallel drawing processing and speed up drawing processing while suppressing an increase in cost due to an increase in hardware resources.

(Modification 1)
Next, a first modification of the present invention will be described. In the band height setting process in the above-described embodiment, the height of the block is set by equally dividing the block height based on the number of parallel operations of the drawing processing unit. On the other hand, in this modification, in addition to the number of parallel operations of the drawing processing units, the specification difference (performance difference) of each drawing processing unit is also considered, and the block height is weighted and the band height is set. .

  FIG. 10 shows a flowchart of the band height setting process in this modification. The processing shown in the flowchart of FIG. 10 is implemented by the CPU 108 executing a program developed in the RAM 109 of the image forming apparatus 100. The band height setting process according to the present embodiment is a modification of the band height setting process described with reference to the flowchart of FIG. Only the differences are described. Specifically, steps S1001 to S1006 in FIG. 10 are the same as steps S701 to S706 in FIG.

  In the band height setting process according to the present embodiment, when the number of parallel operations of the drawing processing unit 114 is 2 or more, in step S1007, the CPU 108 acquires specification difference information of each drawing processing unit that operates in parallel. The specification difference information is, for example, the ratio of the operating frequency of the second drawing processing unit 402 to the first drawing processing unit 401, and the number of processor cores of the second drawing processing unit 402 with respect to the first drawing processing unit 401, for example. And the ratio of the number of modules.

  Next, in step S1008, the CPU 108 determines whether there is a specification difference in each drawing processing unit based on the acquired specification difference information. If the ratio of the specification difference is 1: 1 and there is no specification difference, the process proceeds to steps S1009 to S1010, and the height of the band is divided by equally dividing the block height in the same manner as in steps S707 to S708 in FIG. Is calculated and set in the internal register of the drawing processing unit 114. On the other hand, if the ratio of the specification difference is, for example, 3: 1 and there is a specification difference, the process proceeds to step S1011.

  In step S <b> 1011, the CPU 108 performs weighted division of the height of a single block based on the specification difference information of the drawing processing units that operate in parallel. For example, as described above, when the ratio of the specification difference is 3: 1 (the number of parallel operations is 2), the number of parallel operations and the specification difference are similar to those of the band 11 and the band 12 illustrated in FIG. According to the ratio, the height of a single block is weighted and divided. For example, when the height of a single block is 8 pixels, the drawing area height of the first drawing processing unit 401 is divided into 6 pixels, and the drawing area height of the second drawing processing unit 402 is divided into 2 pixels. To do.

  In step S1012, the value calculated by weighted division is set as the band height in the internal register of the drawing processing unit 114.

  As described above, according to this modification, in view of the number of parallel operations and performance differences (specification differences) of a plurality of drawing processing units operating in parallel in the drawing processing unit 114, the difference in drawing speed is reduced. By dividing the drawing area into two, the efficiency of parallel operation can be increased. That is, when performing drawing processing in units of bands in parallel, by reducing the waiting time that occurs when the rate is determined to be the slower of the processing speeds of the first drawing processing unit 401 and the second drawing processing unit 402, The drawing process can be speeded up.

(Modification 2)
Next, a second modification of the present invention will be described. In this modification, as a method of equally dividing the height of a single block by the number of parallel operations, as shown in FIG. 9D or FIG. Adopt the method to do.

  In the example shown in FIG. 9D, the number of parallel operations of the drawing processing unit is 2, whereas the height of a single block is equally divided by 4 which is twice that number. Then, control is performed so that the first drawing processing unit 401 draws the bands 11 and 13 and the second drawing processing unit 402 draws the bands 12 and 14.

  Alternatively, in the example shown in FIG. 9E, the number of parallel operations of the drawing processing unit is 2, whereas the height of a single block is equally divided by 8 that is four times that. Then, control is performed so that the first drawing processing unit 401 draws the band 11, the band 13, the band 15, and the band 17, and the second drawing processing unit 402 draws the band 12, the band 14, the band 16, and the band 18.

  As described above, according to the present modification, the drawing object larger than the band height can be arranged so as to extend over a plurality of bands by making the band height smaller. By doing so, it is possible to efficiently perform parallel drawing processing by reducing the difference in drawing speed between the drawing processing units operating in parallel. That is, since the drawing load of the first drawing processing unit 401 and the second drawing processing unit 402 can be made uniform, the waiting time that occurs when the rate is controlled to the later one of the first drawing processing unit 401 and the second drawing processing unit 402 By shortening, drawing processing can be speeded up.

(Modification 3)
Next, a third modification of the present invention will be described.

  FIG. 11 shows an example of the relationship between blocks and bands in a page defined by the present invention. As described above, a block is a rectangular area as a unit when image data in page units is divided and handled in the image forming apparatus, and has, for example, a square tile shape of 32 pixels × 32 pixels. As described above, the band is a rectangular area having a larger number of pixels in the main scanning direction and a smaller number of pixels in the sub-scanning direction than the block.

  The relationship between the block and the band shown in FIG. 11A is a positional relationship in which the block is included in the band, and the height of the block is equal to the height of the band. The relationship between the block and the band shown in FIG. 11B and FIG. 11D is a relationship in which the block and the band are not included in each other, and the height of the block is divided by the height of a plurality of bands. In any of FIGS. 11A, 11B, and 11D, the band width (number of pixels in the main scanning direction) is equal to the page width. In this modification, the relationship between the blocks and the bands shown in FIGS. 11A, 11 </ b> B, and 11 </ b> D may be switched depending on the drawing processing setting of the image forming apparatus 100.

  11 (c) and 11 (e) show the relationship between blocks and bands corresponding to FIG. 11 (b) and FIG. 11 (d), respectively, but FIG. 11 (c) and FIG. 11 (e). Then, the bandwidth is divided in the main scanning direction. Thus, the bandwidth is not limited to the page width, and may be smaller than the page width. In this case, among the divided bands, first, drawing processing is performed on all the bands on the left side, and then drawing processing is performed on the right band, thereby performing drawing processing for each page.

  As described above, according to the present modification, parallel drawing is performed while further reducing the required memory capacity by dividing the band unit in the main scanning direction and reducing the image data to perform drawing processing. Processing can be performed efficiently and drawing processing can be speeded up.

(Other embodiments)
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program. It can also be realized by processing. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

Claims (7)

An image processing apparatus that performs drawing processing by dividing intermediate data generated from drawing data into a plurality of band regions, and generates raster format image data,
A plurality of drawing means for performing drawing processing in parallel with respect to each of a plurality of band-by-band intermediate data corresponding to each of the plurality of band areas, wherein the band area performs main scanning with respect to a predetermined block area; A plurality of drawing means having a large number of pixels in the direction and a small number of pixels in the sub-scanning direction;
Storage means for saving and storing image data in band units generated from the intermediate data when storage capacity is insufficient in the rendering process;
Each of the plurality of drawing units synthesizes the saved image data read from the storage unit with image data generated by performing the drawing process on the intermediate data anew, and outputs the raster format image. An image processing apparatus that generates data.   2. The image according to claim 1, wherein the height of the band area is a value obtained by equally dividing the number of pixels in the sub-scanning direction of the predetermined block area by the number of parallel operations of the plurality of drawing units. Processing equipment.   2. The height of the band area is a value obtained by weighting and dividing the number of pixels in the sub-scanning direction of the predetermined block area based on a specification difference between the plurality of drawing units. An image processing apparatus according to 1.   When the storage unit detects that the reading of the band unit image data by the plurality of drawing units is completed, the storage unit stores block unit image data corresponding to the new band unit image data generated from the intermediate data. The image processing apparatus according to claim 1, wherein start of writing is controlled.   When the storage unit detects that the writing of the block unit image data corresponding to the band unit image data is completed, the storage unit controls the start of reading the band unit image data by the plurality of drawing units. The image processing apparatus according to claim 1.   A program for causing a computer to function as the image processing apparatus according to any one of claims 1 to 5. An image processing method in which intermediate data generated from drawing data is divided into a plurality of band regions and drawing processing is performed, and raster format image data is generated,
A drawing step of performing drawing processing in parallel on each of a plurality of intermediate data in units of bands corresponding to each of the plurality of band areas by a plurality of drawing means, wherein the band areas are arranged in predetermined block areas. On the other hand, a drawing process in which the number of pixels in the main scanning direction is large and the number of pixels in the sub-scanning direction is small;
A storage step of saving image data in band units generated from the intermediate data and storing the image data in a storage means when storage capacity is insufficient in the drawing process,
The drawing step generates the raster format image data by combining the saved image data read from the storage unit with the image data generated by newly performing the drawing process on the intermediate data. An image processing method comprising the step of:

Images